Thrust imbalances at high altitudes

[quasarrgames]

Why do turbojets get massive thrust imbalances at high altitudes? Does it happen to planes in real life? Those are my questions.

Also, i've seen people counteract this by simply adjusting the thrust limit for the engines. When i do it, it's extremely tedious and requires semi-decent reaction time, which i don't have. I'm just wondering why someone couldn't (or more correctly hasn't) made a computer part that automatically does it for you? What do you think?

[qromodynmc]

Well,Once in a time,I was using Mechjeb for this,It is automatically adjusting thrust to incoming air,and prevents flameout..

[Red Iron Crown]

It has to do with one engine starving for intake air before the other, a consequence of the way the intake air calculation is handled by KSP. Real planes do not suffer from this as intakes are by necessity assigned to one engine or the other.

More detail on how to design to reduce (but not eliminate!) the effect here, with more detail here.

[Kasuha]

I wonder what role does the poll play here - if you wanted to come up with a suggestion, you should use the Suggestions forum. The game development is not driven by public polls so I believe developers will not really pay attention to it.

The behavior is of course artefact of game implementation (real turbojets have no chance of working at such altitudes) and mechanics how it works and how to prevent the problem are known.

[Wanderfound]

Why do turbojets get massive thrust imbalances at high altitudes? Does it happen to planes in real life? Those are my questions.

Also, i've seen people counteract this by simply adjusting the thrust limit for the engines. When i do it, it's extremely tedious and requires semi-decent reaction time, which i don't have. I'm just wondering why someone couldn't (or more correctly hasn't) made a computer part that automatically does it for you? What do you think?

Yes, it does happen in real life. It was the major safety issue affecting the SR-71. As soon as the engines start to sputter from lack of oxygen, a bit of yaw was induced which in turn leads to a positive feedback loop as the intake of the lower-thrust engine rotates away from the airstream.

There are an assortment of ways to deal with it. Placing your intakes and engines in the right order can reduce thrust asymmetry; using RCS or Vernors for vectored thrust can counteract the yaw; building a plane with an odd number of engines and shutting down all but the central jet at extreme altitudes can eliminate asymmetry altogether. And, as you mention, keeping an eye on the situation and throttling down or switching to rocketry at the right time can also manage the issue.

Hypersonic aircraft are supposed to be a bit tricky to fly; it's just part of the deal.

Klockheed Martian has a flameout detector part that can shut down a pair of engines as soon as one begins to sputter, BTW.

[Levelord]

I've been told the jets in KSP are not the same as real life jets. With that said, they are quirky and fun enough for me to enjoy regardless.

The jet flameouts at high altitudes are quite easy to handle once you learn to fly at a very gradual climb above 25km. Most people will open up their engine menus to see which engine is beginning to flame out, but I take a fairly simple approach and look at the yaw meter to see if my plane is trying to compensate for any loss of thrust in an engine.

Then you simply need to throttle back slowly. Trying to counteract flameouts is difficult when the rate of ascent is very high (flameouts are sudden this way), so the more gradual climb you have, the slower the flameouts will take to appear, giving you ample time to react.

[Wanderfound]

The poll needs a "meh" option, BTW.

I don't want an auto thrust balancer for myself, but I'm not bothered if anyone else chooses to use such a thing. Managing issues like this is fun for me, but my fun ain't necessarily the same as your fun.

[Stratzenblitz75]

Just to let you all know, there IS a good reason other than ease of use to place your intakes/jet engines individually. By having your intake air distributed equally across all your jet engines you get a more powerful aircraft that is able to run off of jet power much longer. Take a look at this test I performed:

I built one craft by placing jet engines and intakes in this order:

Jet engine

Intake

Intake

Jet engine

Intake

Intake

...

Here's how it did carrying a 41 ton payload:

Javascript is disabled. View full album

Then I removed all the jet engines and intakes and placed all the jet engines first, then all the intakes.

Here's how it did carrying the same payload following the same flight profile:

Javascript is disabled. View full album

While the first craft almost made orbit, this one didn't even get close as it kept falling back down and losing speed to the lower atmosphere every time I tried to ascend.

[O-Doc]

I've been told the jets in KSP are not the same as real life jets. With that said, they are quirky and fun enough for me to enjoy regardless.

The jet flameouts at high altitudes are quite easy to handle once you learn to fly at a very gradual climb above 25km. Most people will open up their engine menus to see which engine is beginning to flame out, but I take a fairly simple approach and look at the yaw meter to see if my plane is trying to compensate for any loss of thrust in an engine.

http://i.imgur.com/w64KhCn.png

Then you simply need to throttle back slowly. Trying to counteract flameouts is difficult when the rate of ascent is very high (flameouts are sudden this way), so the more gradual climb you have, the slower the flameouts will take to appear, giving you ample time to react.

This. I don't think the engine flameout is even an issue if you're paying attention to your ascent.

[John FX]

Just to let you all know, there IS a good reason other than ease of use to place your intakes/jet engines individually. By having your intake air distributed equally across all your jet engines you get a more powerful aircraft that is able to run off of jet power much longer. Take a look at this test I performed:

I built one craft by placing jet engines and intakes in this order:

Jet engine

Intake

Intake

Jet engine

Intake

Intake

...

Here's how it did carrying a 41 ton payload:

http://imgur.com/a/nzQpx

Then I removed all the jet engines and intakes and placed all the jet engines first, then all the intakes.

Here's how it did carrying the same payload following the same flight profile:

http://imgur.com/a/G8CaJ

While the first craft almost made orbit, this one didn't even get close as it kept falling back down and losing speed to the lower atmosphere every time I tried to ascend.

very interesting. Glad to see there is a way to not suffer from asymmetrical engine sputter.

[Red Iron Crown]

This. I don't think the engine flameout is even an issue if you're paying attention to your ascent.

Even if you manage the throttle well and don't flameout, unevenly distributed intakes will mean you have to manage the throttle sooner.

[O-Doc]

Even if you manage the throttle well and don't flameout, unevenly distributed intakes will mean you have to manage the throttle sooner.

Air = Thrust

It doesn't matter how you distribute your intakes.

[Stratzenblitz75]

Air = Thrust

It doesn't matter how you distribute your intakes.

Yes, it does matter how you distribute your intakes. Have you seen the experiment that I've performed to test this?

The craft with the evenly distributed intakes performs FAR better than the one that doesn't.

[Wanderfound]

I'm still a partisan of the engine shutdown approach, though. Regardless of build order, one engine fed by x number of intakes will always flame out at a higher altitude than more than one engine being fed by x number of intakes.

If you want to go high without burning oxidiser, you want a single air-breather on the centreline and enough lift and pitch authority to maintain altitude with minimal thrust in thin air.

[Cirocco]

Just to let you all know, there IS a good reason other than ease of use to place your intakes/jet engines individually. By having your intake air distributed equally across all your jet engines you get a more powerful aircraft that is able to run off of jet power much longer. Take a look at this test I performed:

I built one craft by placing jet engines and intakes in this order:

Jet engine

Intake

Intake

Jet engine

Intake

Intake

...

Here's how it did carrying a 41 ton payload:

-snip-

Then I removed all the jet engines and intakes and placed all the jet engines first, then all the intakes.

Here's how it did carrying the same payload following the same flight profile:

-snip-

While the first craft almost made orbit, this one didn't even get close as it kept falling back down and losing speed to the lower atmosphere every time I tried to ascend.

Wait, the order in which you place engines and air intakes in the SPH matters?! This would explain SO much! And it would make my SSTO's about 1000% more effective! I must test this when I get home, you sir or madam, may just have solved something that has been bugging me for almost a year!

[Red Iron Crown]

Air = Thrust

It doesn't matter how you distribute your intakes.

For a single engine, that is true. For multiple engines, improperly distributed intakes means one engine reaches conditions where it must be throttled long before the other; the engine getting the lion's share of the air cannot use it all because it's being throttled along with the other engine. Evenly distributed intakes allow the engines to reach their air limit at close to the same time (there's still asymmetric thrust loss, it just starts at higher altitude).

Simply put, craft with evenly distributed intakes perform better. While it is certainly possible to get satisfactory performance without allocating intakes properly, there is some performance loss when doing so.

Take a look at Stratzenblitz75's experiment that he linked, or the experiments in Kasuha's Fuel Flow Rules tutorial; the order of placement of intakes and engines is a large factor in performance.

[Wanderfound]

For a single engine, that is true. For multiple engines, improperly distributed intakes means one engine reaches conditions where it must be throttled long before the other; the engine getting the lion's share of the air cannot use it all because it's being throttled along with the other engine. Evenly distributed intakes allow the engines to reach their air limit at close to the same time (there's still asymmetric thrust loss, it just starts at higher altitude).

Simply put, craft with evenly distributed intakes perform better. While it is certainly possible to get satisfactory performance without allocating intakes properly, there is some performance loss when doing so.

Take a look at Stratzenblitz75's experiment that he linked, or the experiments in Kasuha's Fuel Flow Rules tutorial; the order of placement of intakes and engines is a large factor in performance.

Not disagreeing with any of your points except for: by the time you get to throttle down level, you should already be starting to shut engines off entirely if you've got more than one. The only one I use the throttle down trick on is the final engine. One engine working well beats two engines choking. If it's a big 'un, toggle in pairs; most of my hefty cargo stuff is set up so I can choose from 7/5/3/1 engines at the flick of a button.

[Red Iron Crown]

Not disagreeing with any of your points except for: by the time you get to throttle down level, you should already be starting to shut engines off entirely if you've got more than one. The only one I use the throttle down trick on is the final engine. One engine working well beats two engines choking. If it's a big 'un, toggle in pairs; most of my hefty cargo stuff is set up so I can choose from 7/5/3/1 engines at the flick of a button.

That is true for odd numbers of engines but not even ones, for an even number to be balanced you eventually get down to two engines, which is a popular configuration with the SP+ "pants" part mounted centerline.

I've taken a page from your book on this as my spaceplanes grow larger, odd numbers of engines are easier to manage well by shutting them down in pairs.

[Kasuha]

Not disagreeing with any of your points except for: by the time you get to throttle down level, you should already be starting to shut engines off entirely if you've got more than one. The only one I use the throttle down trick on is the final engine. One engine working well beats two engines choking. If it's a big 'un, toggle in pairs; most of my hefty cargo stuff is set up so I can choose from 7/5/3/1 engines at the flick of a button.

I had my share of experimenting with that when I was getting this beast to orbit. In fact, i discovered intake air distribution rules through my effort to get it to orbit and I can tell you, the difference is HUGE. Although it had three intakes per engine and I was shutting engines in pairs as soon as they started flaming out and experimented with throttling down and disabling engines, there was no way to get it to orbit until I discovered those rules and interleaved them properly.

Most important fact is, you get significantly better thrust at full throttle from more air-starved engines than at any throttle from less non-air-starved engines.

[Wanderfound]

That is true for odd numbers of engines but not even ones, for an even number to be balanced you eventually get down to two engines, which is a popular configuration with the SP+ "pants" part mounted centerline.

I've taken a page from your book on this as my spaceplanes grow larger, odd numbers of engines are easier to manage well by shutting them down in pairs.

That guideline actually goes away once you get big enough, BTW. The Wedgetail has a pair of turbos on the centre line, but it has so much inertia that thrust asymmetry only matters if its happening out on the flanks.

[Teoman2]

It is so annoying when you build your Inter-Planetary Plane that took you an hour, then you get jet engine turn and everything goes haywire. Then you check the forums to see what happened and also cause you are angry at your own stupidity. Luckily that never happened to me because I don't know how to do that cool stuff...

[O-Doc]

Take a look at Stratzenblitz75's experiment that he linked, or the experiments in Kasuha's Fuel Flow Rules tutorial; the order of placement of intakes and engines is a large factor in performance.

I don't see anything unusual in those Imgur albums. The only difference between his high TWR ascent and my high TWR ascents is the fact that he doesn't have to power down a touch between 26km and 30km as my engines will decrease power assymmetrically through that part of the engine power curve. At 30km I power my engines back up to full throttle at 30km to 37km as well. This is not some mysterious air-intake exploit.

[Levelord]

I suppose if you really want to avoid thrust imbalances you could engineer it out of your craft.

At 1:15 you can see the plane flying with just one engine.

[Stratzenblitz75]

I don't see anything unusual in those Imgur albums. The only difference between his high TWR ascent and my high TWR ascents is the fact that he doesn't have to power down a touch between 26km and 30km as my engines will decrease power asymmetrically through that part of the engine power curve. At 30km I power my engines back up to full throttle at 30km to 37km as well. This is not some mysterious air-intake exploit.

You might want to look at the experiment again; I'm comparing the performance of two crafts which have the same number of intakes, engines, the same payload and who follow the same flight plan. The ONLY difference between them is the way their intakes are set up. The experiment clearly shows that the craft with the equally balanced intakes (the first Imbur album) performs FAR better as it almost left the atmosphere. The craft with the unbalanced intakes (the second Imgur album) barely made it above 30 Km. Kasuha talks about this effect here and demonstrates the intake-air flow logic in his thread.

Also, you state that this is "not some mysterious air-intake exploit", yet you don't provide any solid evidence to support your claim. How many intakes per engine did you use for your ascent? What was your TWR? How much payload are you lifting? Did you compare the performance of crafts with equally balanced intakes vs. unequal intakes? Without this information, I'm afraid I can't take your claim seriously.

[magnemoe]

I don't see anything unusual in those Imgur albums. The only difference between his high TWR ascent and my high TWR ascents is the fact that he doesn't have to power down a touch between 26km and 30km as my engines will decrease power assymmetrically through that part of the engine power curve. At 30km I power my engines back up to full throttle at 30km to 37km as well. This is not some mysterious air-intake exploit.

The point is to avoid one engine to get all the air something who happens if you first put on the engines and then the intakes.

Using mechjeb jet flame out protection would give both the same performance, however it simply limit trust to the air you have, the other is to use an setup with an core engine and shut down all but it at high attitude.